The potential of a combined cell‐based therapy and rehabilitation approach for stroke recovery Authors: Abdulhameed Bakreen, Jukka Jolkkonen Published: 2023 ABSTRACT AbstractActivation of neuroprotective and particularly later neurorestorative mechanisms after stroke attempts to restore or compensate for lost functions. This potentially opens a wide window for restorative therapies to promote brain repair and improve long‐term functional recovery. Although extensively demonstrated in the preclinical setting, the efficacy of cell‐based therapies in stroke patients has been modest at best, if any at all. Translational failure may be due to the ineffective survival and integration of transplanted cells in pro‐death stroke microenvironments that are not conducive for the structural reconstruction of damaged brain tissue and repair‐related network reorganization. Optimal systemic delivery, timing, cell product, and dose remain open as well. Fortunately, a better understanding of the brain plasticity mechanisms underlying stroke recovery has ushered in a combination approach of cell‐based therapy and rehabilitation that is aimed at achieving additive, synergistic, or even maximal beneficial effects. This novel combination therapy is not only targeted at promoting exogenous and endogenous cell survival and augmenting stand‐alone restorative mechanisms but also at utilizing rehabilitation to facilitate a graft–host structural and functional integration and plasticity that would effectively remodel stroke tissue and restitute lost functions. This review presents an overview of the combination of cell‐based therapy and experimental rehabilitation in stroke models. It also discusses associated shortcomings as well as proposes strategies to address them and help facilitate the advancement of this combination approach.Keywords: brain regeneration, cell‐based therapy, combination therapy, neuroprotection, neurorepair, rehabilitation, stroke INTRODUCTION Stroke is a leading cause of long‐lasting adult disability and a global burden to patients and healthcare. More than half of patients are left with permanent impairments, and there is no effective treatment to enhance functional recovery. Much hope has been placed on new biological treatment approaches to bring long‐awaited breakthroughs in the stroke recovery field. The endogenous repair mechanisms taking place in the recovery process after stroke are activated in a delayed manner. 1 This opens an extended window for restorative strategies to induce therapeutic plasticity and facilitate a long‐term overall behavioral improvement by maximizing functional restitution in viable brain areas and promoting compensation. 2 The potential of cell‐based therapy has been comprehensively demonstrated in animal stroke models, as different treatment protocols involving several types of cells from various sources have resulted in behavioral improvements associated with attenuated infarct size and enhanced tissue repair. 3 , 4 Yet, the efficacy of cell‐based therapies in the recovery of stroke patients has been modest at best, notwithstanding an increasingly recognized safety. 3 , 4 , 5 Cell‐based therapy is an attractive and promising strategy for the enhancement of structural brain repair or restoration of lost neurovascularity, and the beneficial modulation of the immune response after stroke. 6 , 7 A major factor limiting therapeutic efficacy, however, is the poor survival, migration, homing, or integration of cells after stereotactic transplantation due to the pro‐death microenvironment in the poststroke brain. The evolving neuronal injury and disrupted neural networks associated with stroke are characterized by hostile proinflammatory responses, activated primary and secondary cell death pathways, and inadequate trophic support. 6 , 8 This nonconducive microenvironment not only limits the survival of transplanted cells or the ability of a cell‐based therapy to confer neuroprotection and promote neurovascular repair but also inhibits the recruitment of endogenous neural stem/progenitor cells (NSPCs) and the establishment of regenerative niches. 4 , 9 , 10 , 11 Another major factor limiting therapeutic effects on functional outcome is the transplantation of immature exogenous cells in integrated neural networks of mature endogenous cells. Although key in recapitulating normal brain development, there is the challenge of immature cell grafts forming neurovascularity with mature host brain cells in lost, disrupted, or even spared networks after stroke, while still conferring the general therapeutic effect of cell transplantation. 12 Consequently, an optimized graft–host integration that not only structurally reconstructs lost tissue and functionally repairs lost circuits but also significantly enhances network reorganization towards the maximization of stroke recovery remains elusive. 12 , 13 To overcome these challenges, a combination of cell‐based therapy and rehabilitation that is targeted at promoting cell survival and integration and enhancing stroke tissue remodeling is emerging in the preclinical setting as an approach to achieve additive, synergistic, or maximized therapeutic effect on behavioral recovery. 13 , 14 The therapeutic potential of this combination approach, designed to better enhance the regenerative features of perilesional and remote microenvironments and further promote adaptive brain plasticity, is multifaceted and involves complex and interwoven mechanisms. As the cell‐based therapy enhances endogenous neuroprotection as well as structural and functional brain repair, the add‐on rehabilitation can promote cell survival and differentiation, facilitate a more conducive regenerative niche, and augment plasticity processes, including network reorganization. 13 , 15 At the same time, the rehabilitation may also (i) train or teach the exogenous cells to learn endogenous biological signals or cues in the brain towards better graft–host structural integration and functional connectivity and (ii) stabilize or reinforce the reconstructed networks that are beneficial in addition to pruning the nonuse or dysfunctional ones. 12 Ultimately, the transplanted cells or cell‐derived biomolecules (priming therapy) may serve as the building blocks in the creation of synergistic plasticity‐related regenerative microenvironments that, aided by rehabilitation (consolidation therapy), direct facilitated repair processes and newly formed circuitries (spontaneously‐, cell‐, and/or rehabilitation‐induced) to result in a more robust behavioral recovery. 14 , 16 Experimental studies show that repetitive training or housing in an enriched environment (EE) upregulates signaling molecules, promotes adaptive brain plasticity, and mediates the molecular pathways involved in activity‐ or experience‐induced learning and acquisition of skills. 17 Indeed, voluntary exercise, forced physical training, and EE are validated experimental rehabilitation strategies for stroke that have proven their efficacy in promoting neuroprotection and neurorestoration, and improving functional outcomes. 18 , 19 Voluntary exercise and forced physical training improve cognitive and motor outcomes in stroke animals by enhancing trophic factor secretion, promoting neuroprotection, and boosting neurogenesis. 20 , 21 , 22 Similarly, EE adds a multimodal and nonspecific configuration of physical, sensorimotor, cognitive, or social stimulation into a housing environment and exerts positive effects on angiogenesis, neurogenesis, axonal sprouting, and network remodeling. 23 , 24 Recent studies on the mechanisms underlying brain plasticity, motor learning, and stroke recovery have led to a better understanding and a deeper dive into possible strategies by which rehabilitation may be used in conjunction with cell‐based therapy to achieve better functional outcomes. Emerging evidence suggests that effectively combining priming and consolidation therapies and maximally harnessing all possible beneficial effects holds significant promise and may be the future of the long‐term management of stroke patients. 14 Current standing of cell‐based therapy or rehabilitation as a stand‐alone restorative strategy for stroke is already extensively reviewed in the literature. 3 , 4 , 18 , 19 The subsequent sections will therefore focus on the promise of combined cell‐based therapy and experimental rehabilitation to additively or synergistically promote regenerative microenvironments, further enhance neurovascular function, and better improve behavioral recovery in experimental stroke animals. Possible combinatorial therapeutic interactions will be summarized (Figure 1) and related to reported behavioral outcomes in the literature (Table 1). Finally, considerations that could be crucial in resolving the challenges associated with this combination therapy will be discussed. Schematic illustration of possible mechanisms and interactions underlying the efficacy of a combined cell‐based therapy and experimental rehabilitation approach for stroke rodents. Cell‐based therapy exerts stand‐alone beneficial effects on functional recovery in stroke rodents by activating multiple mechanisms involved in neuroprotection and neurorestoration. These mechanisms include the enhancement of trophic effect, promotion of cell survival, recruitment, or replacement, as well as immunomodulation. Experimental rehabilitation exerts add‐on beneficial effects by promoting the survival, migration, proliferation, or integration of exogenous cells, which in turn boosts the therapeutic potential of the transplanted cells. Experimental rehabilitation also trains the recruited endogenous neural stem cells/progenitors as well as the surviving and mature neural tissue involved in the formation of new connections. Furthermore, experimental rehabilitation influences the multiple neuroprotective and neuroreparative mechanisms activated by the cell‐based therapy, altogether resulting in an adaptive structural and functional brain repair. In an ascending order, the stand‐alone therapies can interact with one another to produce an overlapping, antagonistic, sub‐additive, additive, or synergistic effect on the functional outcome. The interaction can also lead to a maximized therapeutic effect on functional recovery. Created with BioRender.com. Table 1Studies that have employed the combination of cell‐based therapy and experimental rehabilitation in stroke models.ModelCombination strategyCombination effect & main functional outcomesUnderlying mechanismsReferencesRat tMCAOIC mouse SVZ‐NSCs at 7 d + EE at 8 d for 1 month Undetermined combination effect. Improvement in sensorimotor recovery is transient and only better than vehicle control Increased migration of transplanted cells, and enhanced proliferation of endogenous SVZ neuroblasts[25]Rat tMCAOTreadmill training at 24 h for 1 month + IC human ESC‐derived neuronal progenitor cells at 7 dAntagonistic effect of cell transplants on treadmill training in the improvement of gross neurological functionThe decrease in infarct size was antagonized by complications from the invasive cell infusion, that is, increase in neural cell loss and detrimental reactive gliosis[26]Rat tMCAOIC mouse SVZ‐NSCs at 7 d + EE at 8 d for 3 monthsOverlapping/no combination effect and no stand‐alone improvement in sensorimotor recoveryAbout 99% of cell transplants dead by 2 months, enhanced activation of microglia[27]Rat pMCAOIC human ESC‐NPCs at 7 d + EE at 7 d for 2 months Overlapping/no combination effect. Improvement in the recovery of sensorimotor function is minimal and not better than the stand‐alone treatments. No improvement in the recovery of skilled motor function after all treatments Poor survival of the transplanted cells[28]Neonatal mouse chronic hypoxia‐ischemiaIC human ADMSCs at 35 d + EE at 35 d for 2 monthsSynergistic effects on the improvements in sensorimotor and motor recoveryPromoted endogenous neurogenesis and angiogenesis, enhanced activation of astrocytes, and increased levels of FGF‐2, Ang‐1, Ang‐2, VCAM‐1, MMP‐2[29, 30]Rat tMCAOIV allogenic rat BMMSCs at 24 h + treadmill training at 24 h for 2 weeksTreadmill training has a synergistic effect on the transplanted cells in the improvement of gross neurological functionSuppressed apoptosis via upregulated Bcl‐2 & survivin levels[31]Rat pMCAOIV allogenic rat BMMSCs at 6 h + treadmill training at 24 h for 5 weeksSynergistic effect on the improvement in sensorimotor recoveryAttenuated infarct size, enhanced synaptogenesis, and thicker corpus callosum[32]Mouse tMCAOICV allogenic mouse BMMSCs at 24 h + treadmill training at 24 h for 6 days Treadmill training has a synergistic effect on the cell transplants in the improvement of gross neurological and motor functions in young mice. Overlapping/no combination effect in the middle‐aged mice as well as no stand‐alone effect on the improvement of gross neurological and motor functions In the young mice: attenuated infarct size, suppressed apoptosis, and enhanced proliferation in the SVZ. In the middle‐aged mice: increased apoptosis [33, 34]Rat pMCAOIV human ADMSCs at 2 d + EE at 2 d for 6 weeksAdditive effect on sensorimotor improvementNo effect on angiogenesis or altered microglia/astrocyte number in the perilesional cortex or neurogenesis in the SVZ. No difference in infarct or glial scar volume[35, 36]Mouse pMCAOIV allogenic mouse BMMNCs at 4 weeks + intermittent physical and cognitive training at 3 weeks for 9 weeksUndetermined combination effect. Improvement in the recovery of sensorimotor, motor, or cognitive function when compared to training aloneUpregulation of genes involved in brain metabolism (energy source transporters) in both hemispheres. No difference in brain weight or the volumes of both hemispheres[37, 38]Rat tMCAOIC human NSCs at 2 weeks + treadmill training at 2 weeks for 10 weeksTreadmill training has (i) a subadditive/overlapping effect on the transplanted cells in the recovery of sensory and sensorimotor functions and (ii) an antagonistic effect on the transplanted cells in motor recoveryEnhanced structural and functional connectivity in primary somatosensory, motor, and striatal areas. Promotion of angiogenesis. Increase in the levels of BDNF, IGF‐1, GDF‐10, and Klotho. Decrease in IL‐6 levels. No effect on the survival, migration, or differentiation of grafted cells[39]Rat tMCAOIV allogenic rat BMMSC‐exosomes at 24 h + treadmill training at 24 h for 2 weeksSynergistic effect on the improvement of gross neurological functionInhibited apoptosis, decreased infarct size, increased synaptogenesis, enhanced axonal regeneration, activated JNK1/c‐Jun pathway[40]Rat photothrombosisIC human BMMSCs at 24 h + treadmill training at 24 h for 4 weeksAdditive effects on the improvement of sensorimotor and motor functionsPromoted viability or survival of cell transplants as well as a possible increase in their proliferative activity[41]Rat tMCAOIC human ESC‐NPCs at 3 d + treadmill training at 7 d for 4 weeksAdditive to synergistic effect of treadmill training on cell transplants in the improvement of gross neurological, sensorimotor, and motor functionsReduced infarct size, increased neuronal differentiation, maturation, and integration of grafted cells, enhanced axonal density and angiogenesis, higher levels of BDNF, VEGF, IGF‐1, GDNF, NGF, and NT‐3, as well as upregulated genes of regeneration, neuroprotection, repair, and recovery[42]Rat tMCAOVoluntary exercise at −7 d for 3 weeks + IC human BMMSCs at 24 hOverlapping/no combination effect. Improvement in gross neurological recovery not better than the stand‐alone treatments. No improvement in the recovery of sensorimotor function after all treatmentsDecreased infarct size, increased angiogenesis, enhanced SVZ and SGZ neurogenesis, and higher BDNF and VEGF levels. >99.9% of cell transplants dead at day 3[43]Abbreviations: ADMSCs, adipose tissue‐derived mesenchymal stem cells; BDNF, brain‐derived neurotrophic factor; BMMNCs, bone marrow‐derived mononuclear cells; BMMSCs, bone marrow‐derived mesenchymal stem cells; EE, enriched environment; ESC, embryonic stem cell; IC, intracerebral; ICV, intrathecal/intracerebroventricular; IGF‐1, insulin‐like growth factor 1; IV, intravenous; GDF‐10, growth differentiation factor 10; GDNF, glial cell line‐derived neurotrophic factor; mNSS, modified neurological severity score; NGF, nerve growth factor; NT‐3, neurotrophin‐3; NPCs, neural progenitor/precursor cells; NSCs, neural stem cells; pMCAO, permanent middle cerebral artery occlusion; SGZ, subgranular zone; SVZ, subventricular zone; tMCAO, transient middle cerebral artery occlusion; VEGF, vascular endothelial growth factor (VEGF). Studies that have employed the combination of cell‐based therapy and experimental rehabilitation in stroke models. Undetermined combination effect. Improvement in sensorimotor recovery is transient and only better than vehicle control Overlapping/no combination effect. Improvement in the recovery of sensorimotor function is minimal and not better than the stand‐alone treatments. No improvement in the recovery of skilled motor function after all treatments Treadmill training has a synergistic effect on the cell transplants in the improvement of gross neurological and motor functions in young mice. Overlapping/no combination effect in the middle‐aged mice as well as no stand‐alone effect on the improvement of gross neurological and motor functions In the young mice: attenuated infarct size, suppressed apoptosis, and enhanced proliferation in the SVZ. In the middle‐aged mice: increased apoptosis Abbreviations: ADMSCs, adipose tissue‐derived mesenchymal stem cells; BDNF, brain‐derived neurotrophic factor; BMMNCs, bone marrow‐derived mononuclear cells; BMMSCs, bone marrow‐derived mesenchymal stem cells; EE, enriched environment; ESC, embryonic stem cell; IC, intracerebral; ICV, intrathecal/intracerebroventricular; IGF‐1, insulin‐like growth factor 1; IV, intravenous; GDF‐10, growth differentiation factor 10; GDNF, glial cell line‐derived neurotrophic factor; mNSS, modified neurological severity score; NGF, nerve growth factor; NT‐3, neurotrophin‐3; NPCs, neural progenitor/precursor cells; NSCs, neural stem cells; pMCAO, permanent middle cerebral artery occlusion; SGZ, subgranular zone; SVZ, subventricular zone; tMCAO, transient middle cerebral artery occlusion; VEGF, vascular endothelial growth factor (VEGF). NEUROPROTECTIVE AND NEUROREPARATIVE EFFECTS OF COMBINED CELL‐BASED THERAPY AND REHABILITATION IN STROKE RODENTS Through their cell replacement, cell‐to‐cell interaction, paracrine, or immunomodulatory capabilities, cell‐based therapies provide beneficial effects on the stroke brain by directly and/or indirectly promoting multiple neuroprotective and neurorestorative mechanisms in perilesional and remote regions. 7 , 44 However, the ability of cell transplants to survive, adhere, migrate/transmigrate, home, or differentiate into neural lineage within the local microenvironment influences efficacy. The ability of transplanted cells to locally or systemically secrete therapeutic biomolecules and cargos, promote endogenous cell recruitment, or enhance endogenous repair mechanisms is also very important. 4 , 9 Therefore, a rehabilitation‐induced activation of neuroprotective and neurorestorative mechanisms sounds like a logical approach to promote cell survival and enhance therapeutic potency after stereotactic or systemic transplantation. This may in turn help establish a plasticity‐related regenerative host brain tissue or potentiate systemic effects towards the maximization of neurovascular repair. Next, we will give some examples of mechanisms and pathways that can be promoted in transplanted cells to enhance their survival. The same survival or neuroprotective mechanisms can be potentiated by rehabilitation. Indeed, the survival of exogenous cells in the stroke brain after intra‐arterial (IA) transplantation is associated with an increase in their Miro1 expression, resulting in greater mitochondrial transfer‐mediated cell viability and ATP production. 45 Rehabilitation can be added to further confer neuroprotective effects on exogenous and endogenous cells, including a decrease in reactive oxygen species generation and apoptosis. In fact, forced physical training after intravenous (IV) cell transplantation in stroke rats inhibits apoptosis in the ischemic penumbra and promotes the survival of endogenous and exogenous cells via Bcl‐2 and survivin‐mediated antiapoptotic pathways. 31 After IV transplantation, the migration, proliferation, and neuronal differentiation of modified exogenous cells overexpressing bFGF are also promoted in the stroke brain, 46 wherein neuroprotective and neurorestorative effects can be further promoted by rehabilitation via increased bFGF signaling. Forced physical training can also activate the stromal cell‐derived factor‐1α/cysteine‐X‐cysteine chemokine receptor 4 (SDF‐1α/CXCR4) pathway in the stroke brain after stereotactic or systemic cell delivery, potentially enhancing the expressions of chemotactic and immunological factors that promote cell migration or transmigration to the ischemic injury region. 47 , 48 , 49 Rehabilitation can also increase the survival of endogenous and exogenous cells after stereotactic or systemic transplantation by modulating the signaling pathways that dampen poststroke proinflammatory responses, for example, downregulation of nuclear factor‐Kappa B (NF‐kB) or phosphatidylinosito‐3‐kinase/protein‐serine‐threonine kinase (PI3k/Akt) signaling. 6 The neuroprotective and neuroreparative effects induced in the brain after the IV transplantation of extracellular vesicles (EVs) derived from mesenchymal stem cells (MSCs) can be potentiated by forced physical training via increased JNK1/c‐Jun signaling in the perilesional cortex. 40 More, exogenous cell‐derived EVs can interact with peripheral immune cells and resident microglia to mediate the immune response and attenuate proinflammatory responses after stroke. 50 Intracerebral (IC) transplantation of neural stem cells (NSCs) overexpressing interferon‐γ (IFN‐γ) also attenuates proinflammatory responses in the stroke brain and promotes neurogenesis. 51 Concurrently, the proinflammatory IFN‐γ protects the transplanted NSCs and induce their neuronal differentiation. Rehabilitation can be added to further promote these neuroprotective and neurogenic mechanisms. Endogenous NSPCs and transplanted cells that do not differentiate into mature neurons can also secrete trophic factors and further contribute to overall cell survival and brain repair. In addition, EE can enhance the neurogenic potential of endogenous NSPCs or exogenous cells in the stroke brain and promote neurogenesis in the subventricular zone (SVZ) and striatum through NF‐kB‐mediated secretion of IL‐17A from reactive astrocytes. 52 Reactive astrocytes in stroke rodents housed in EE overexpress GFAP, pointing to a mechanism by which both EE and IC cell delivery promote beneficial astrogliosis and the upregulation of brain repair factors after stroke. 29 , 30 , 53 Furthermore, EE can confer immunomodulatory effects by decreasing IL‐1β, an inflammatory cytokine whose levels are increased after stroke. 53 Voluntary exercise induces an increased level of plasticity‐related proteins in the stroke brain, such as brain‐derived neurotrophic factor (BDNF) and vascular endothelial growth factor (VEGF), which endogenous and transplanted cells also respond to. 20 , 43 BDNF and VEGF are neurotrophic factors whose levels are increased after stroke and are responsible for neural survival, neurovascular remodeling, and glial scar modulation. In fact, voluntary exercise and IC cell transplantation in stroke rats upregulate BDNF and VEGF levels and promote perilesional remodeling by boosting angiogenesis and neurogenesis. 43 Endogenous NSPCs in the SVZ also proliferate after stroke in response to EE via BDNF‐mediated increase in Gadd45b. 54 Taken together, a potential additive or synergistic effect on the establishment of a repair‐promoting and recovery‐permissive stroke microenvironment can be achieved after cell‐based therapy and rehabilitation, since both have similar underlying therapeutic mechanisms. Ultimately, the add‐on rehabilitation may (i) help make exogenous cells or their biomolecules more resistant in harsh microenvironments following their transplantation or help foster nonhostile regenerative niches, (ii) enable the modified exogenous cells or their biomolecules to better execute neurovascular repair via cell replacement or paracrine antiapoptotic, anti‐inflammatory, proangiogenic, and proneurogenic effects on endogenous NSCs/NSPCs and mature cells. However, it is important to note that the mechanisms underlying therapeutic efficacy might be different between stereotactic and systemic transplantations. In the case of intracerebral cell transplantation, rehabilitative training might improve cell survival and integration into host microenvironment in addition to the enhancement of brain tissue repair. Systemic cell delivery, on the other hand, activates various brain repair mechanisms in nonspecific ways with or without transplanted cells necessarily entering into the brain parenchyma. In fact, this combination therapy, although still in its infancy in the preclinical stroke setting, has so far yielded conflicting outcomes due to differences in study design. The route of transplantation influences the migration, mechanisms or mode of action, and efficacy of cell‐based therapy. The following sections will therefore review the literature and summarize the roles of transplantation route, cell type, and rehabilitation in relation to behavioral outcome (Table 1). BEHAVIORAL EFFECTS OF COMBINING THE STEREOTACTIC TRANSPLANTATION OF NSCS OR MSCS WITH EXPERIMENTAL REHABILITATION IN STROKE RODENTS NSCs have a high tissue‐specific regeneration potential owing to their ability to proliferate, migrate, and differentiate into all neural cell types within the local environment. 12 So, the stereotactic transplantation of NSCs may be advantageous in the sense that more cells can reach the specific target brain tissue to confer localized effects via cell replacement and trophic factor secretion, which can be further enhanced by the add‐on rehabilitation. Indeed, the IC transplantation of human ESC‐NPCs combined with forced physical training improves gross neurological, sensorimotor, and motor functions in an additive or synergistic fashion 7 weeks after transient MCAO in rats. 42 Long‐term improvements in gross neurological and sensorimotor functions at 14 weeks were also better when the combination therapy was compared to cell treatment only. This underscores the potential of employing rehabilitation to better improve behavioral outcomes after a cell‐based therapy for stroke. Investigation of the underlying therapeutic mechanisms revealed synergistic effects on infarct size reduction, axonal sprouting, angiogenesis, as well as elevation of the levels of BDNF, VEGF, IGF‐1, GDNF, NGF, and NT‐3. Genes that promote regeneration, neuroprotection, brain repair, and functional recovery after stroke were also upregulated. A deeper dive into the underlying mechanisms also showed that the neurotrophic/growth factor‐enriched regenerative microenvironment fostered by the rehabilitative training promoted the survival as well as the neuronal differentiation, maturation, and integration of the cell transplants. Another study shows, however, that the IC transplantation of human NSCs combined with forced physical training only results in a subadditive/overlapping effect on sensory function and sensorimotor outcome 10 weeks after transient MCAO in rats. 39 In fact, grip strength deficit was significantly improved only in the cell‐treated group when all treated groups were compared to the vehicle controls, with no difference found in the effects of the forced physical training alone and the combined treatment. It was therefore stipulated that the training exerted an antagonistic effect on the transplanted cells in relation to the motor outcome. Curiously, growth factor secretion, angiogenesis, as well as structural and functional connectivity in stroke tissue were enhanced in all the treated groups. Nonetheless, rehabilitative training did not affect exogenous cell survival, migration, or integration after transplantation, which might partly explain why the combination treatment did not result in a maximized therapeutic effect on behavioral outcome. Long‐term impairments, counteraction between behavioral testing and spontaneous recovery, and an insufficient dose of rehabilitative training were also put forward as limiting factors. Notably, the disruption of complex neural networks underlying an array of several behavioral tasks is a characteristic feature of stroke. This suggests that further studies are needed to assess how the combination of stereotactic cell transplantation and rehabilitation can be supplemented with behavioral testing to bolster task integration and attain better functional outcomes. The combination of forced physical training with the IC transplantation of human ESC‐derived neuronal progenitor cells can also result in an antagonistic effect on behavioral recovery 24 days after transient MCAO in rats. 26 When all treatments were compared, training alone resulted in the biggest infarct size attenuation and the best improvement in the recovery of gross neurological functions, fortifying the data on the efficacy of rehabilitative training after stroke. In relation to the behavioral outcome, however, the effect of cell transplantation only was not different to that of the combined therapy, indicating that the therapeutic effect of the forced physical training was antagonized by the cell treatment. This antagonistic interaction between the stand‐alone therapies might have been caused, in part, by the invasiveness associated with IC transplantation. Indeed, loss of neural cells, detrimental reactive gliosis, and microcalcification were all reported at the site of transplantation in the basal ganglia. In three separate studies, Hicks and colleagues assessed the medium or long‐term effect of the combination of the IC transplantation of NSCs/NSPCs and EE in a rat transient or permanent MCAO model. 25 , 27 , 28 The first study reported that combined IC transplantation of mouse SVZ‐derived NSCs and EE after transient MCAO in rats only transiently improved the recovery of sensorimotor function at postoperative day seven. This is despite an EE‐induced facilitation of the migration of transplanted cells towards the infarct region. 25 EE also promoted the proliferation of endogenous SVZ neuroblasts at postoperative day 30. The study did not set up an EE‐only group, though, meaning combinatorial interactions between cell transplants and EE, including a possible additive or synergistic effect, could not be confirmed. Using the same combination strategy and stroke model, an EE‐only treatment group and a longer follow‐up period were incorporated in a following study. 27 Interestingly, the long‐term survival and migration of the transplanted cells were not affected by EE. In fact, none of the three treatments improved the recovery of sensorimotor function 30, 60, or 90 days after the induction of stroke. Since most (approximately 99%) of the transplanted cells were dead within 60 days, and cell survival negatively correlated with microglial activation, the cell transplants were possibly rejected by the immune system. In the third and final study by the group, the IC transplantation of human ESC‐derived neural precursor cells and EE were combined after permanent MCAO in rats, and immunosuppressants were instead continually administered with osmotic minipumps. 28 Still, the survival or differentiation of the transplanted cells were not affected by EE. When all treatments were compared, there was no difference in their effects on the recovery of sensorimotor function at postoperative Day 60. Moreover, skilled motor functions did not improve after all treatments, although the combined therapy exerted some improvement in the spontaneous use of the impaired forelimb when compared to the vehicle controls. MSCs possess the ability to secrete therapeutic substances and confer immunomodulatory effects via paracrine and cell‐to‐cell mechanisms in stroke models. They may also migrate/transmigrate to the injured area and seemingly differentiate into the neural lineage. 7 However, their cell replacement capability is questionable as they cannot survive for long periods at the site of transplantation and do not generate mature and functional neurons to replace the lost ones, 55 limitations that could be resolved by an adjuvant rehabilitation. Indeed, voluntary exercise combined with the IC transplantation of human BMMSCs may improve gross neurological but not sensorimotor recovery at postoperative Day 14 in a rat transient MCAO model. 43 The underlying therapeutic mechanisms put forward include the synergistic effects on the attenuation of infarct size, angiogenesis, as well as BDNF‐ and VEGF‐mediated neurogenesis. The low survival (<0.05%) of the transplanted cells at Day 3, however, likely as a result of technical issues, might partly explain the overlapping/limited combinatorial effect on the behavioral outcomes. On the other hand, combined IC transplantation of human BMMSCs and forced physical training improves the recovery of sensorimotor and motor functions in an additive manner 4 weeks after photothrombosis in rats. 41 Even though the effects of forced physical training only and vehicle control were equivalent in relation to the behavioral outcomes, the training promoted exogenous cell viability or survival 7 days after transplantation. The training also likely enhanced the proliferation of the transplanted cells. In addition, IC transplantation of human ADMSCs combined with EE synergistically improves sensorimotor and motor functions 2 months after the induction of chronic hypoxia‐ischemia in neonatal mice. 29 , 30 A high engraftment and neural differentiation of the transplanted cells as well as the promotion of endogenous neurogenesis in the SVZ and striatum were reported as underlying therapeutic mechanisms. FGF‐2 upregulation leading to an increase in the activation of astrocytes was also proposed. 29 A more thorough investigation of the underlying mechanisms in a follow‐up study also showed that a synergistic increase in the levels of the angiogenic factors FGF‐2, Ang‐1, Ang‐2, VCAM‐1, and MMP‐2 was facilitated by enhanced astrocytic activation and endogenous angiogenesis. 30 It should be noted, though, that functional outcome was assessed in 3‐month‐old young mice in which chronic stroke was induced when they were still 7‐day‐old neonates. Also, one of the studies used only three animals in the investigation of the underlying mechanisms. 29 Furthermore, combined intracerebroventricular (ICV) transplantation of allogenic mouse BMMSCs and forced physical training synergistically improves gross neurological and motor recovery in a young mouse transient MCAO model 7 days after operation. 33 The training seemingly boosted the therapeutic efficacy of the cell transplants by facilitating synergistic effects on the suppression of apoptosis, reduction of infarct size, and enhancement of cell proliferation in the SVZ. This further reinforces the potential in the use of rehabilitation as an adjuvant to cell‐based therapy towards better tissue and functional outcomes after stroke. Interestingly, the training did not affect the efficacy of transplanted cells in a subsequent study that also employed the same combination strategy, stroke model, and postsurgery follow‐up period, albeit in mice at middle age. 34 Contrary to the findings from the previous study, the loss of neural cells in the middle‐aged stroke mice was not affected by the training on postoperative Day 7. In fact, there was no effect of cell transplantation alone, training alone, or their combination on the behavioral outcomes, by and large highlighting the significance of age in recovery after stroke. Taken together, stereotactic transplantation in the brain may be technically more invasive and associated with more undesirable effects compared to intravascular transplantation. This might be the reason why the IC transplantation of cells had an overall limited stand‐alone therapeutic effects on behavioral outcomes in these combination strategies, which required the significant add‐on beneficial effects of the adjuvant rehabilitative training. However, the cell type, cell source, and rehabilitation strategy also contributed to the variability in outcomes. Indeed, stereotactic transplantation plus experimental rehabilitation is associated with much improved regenerative microenvironment and tissue repair in the stroke brain. It also seems that forced physical training is more effective than voluntary exercise or EE in promoting exogenous cell survival and integration, though the treatment sequence is also an important factor. Further studies are needed to better understand how different rehabilitation modalities can be exploited to effectively potentiate the specific mechanisms or mode of action underlying the therapeutic effects of various cell types or sources after stereotactic transplantation. BEHAVIORAL EFFECTS OF COMBINING THE SYSTEMIC TRANSPLANTATION OF MSCS WITH EXPERIMENTAL REHABILITATION IN STROKE RODENTS IV transplantation of cells is minimally invasive and maybe the most suitable for extensive lesions, although engraftment in brain tissue is low because of cell accumulation in the filtering or inflammation‐rich internal organs. 56 Rehabilitation can be added to further promote overall beneficial effects on the stroke brain. Combined IV transplantation of allogenic rat BMMSCs and forced physical training has a synergistic therapeutic effect on the recovery of gross neurological functions 2 weeks after transient MCAO in rats. 31 A synergistic increase in the levels of the apoptotic modulators Bcl‐2 and survivin, which led to a considerable inhibition of the apoptosis of endogenous and transplanted cells, was suggested as the underlying mechanisms. In fact, the neuroprotective effects of the stand‐alone therapies were similar, and the training was credited for the significant increase in exogenous cell survival and suppression of overall cell death after the combined therapy. Yet again, this reinforces the notion that rehabilitation may help foster a recovery‐permissive environment for transplanted cells in the stroke brain. Consistently, the same combination strategy also has a synergistic effect on the recovery of sensorimotor functions 5 weeks after permanent MCAO in rats. 32 Herein, synergistic effects on the decrease in infarct size, increase in thickness of the corpus callosum, as well as synaptogenesis were implicated as the underlying therapeutic mechanisms. In a more recent study, combination of the IV transplantation of allogenic rat BMMSC‐derived exosomes and forced physical training also improved gross neurological recovery in a synergistic manner 2 weeks after transient MCAO in rats. 40 Inhibition of apoptosis, reduction of infarct size, as well as the promotion of synaptogenesis and axonal regeneration were proposed as the underlying mechanisms. In addition, the reported antiapoptotic and neuroreparative effects were likely mediated by an activated JNK1/c‐Jun pathway in the perilesional cortex. Our group combined the IV transplantation of human ADMSCs with EE in a rat permanent MCAO model and observed, at the least, additive effects on two different sensorimotor functions 6 weeks after the induction of stroke. 35 Challenges were encountered in the discrimination of add‐on therapeutic effects, though, due to a complex study design and the spontaneous recovery of the animals, perhaps limiting the chances of detecting a potential synergistic effect. Also, there were no stand‐alone and combined treatment effects on the investigated underlying mechanisms, including angiogenesis, neuroinflammation, or astrogliosis in the perilesional cortex 35 , 36 and neurogenesis in the SVZ (data published in a doctoral dissertation). However, the spontaneous recovery or behavioral improvement in parallel with the long‐term alterations in perilesional and SVZ microenvironments, despite the nonexistence of cell transplants in the brain, suggests that repair and recovery mechanisms might have been systemically and beneficially modulated. Finally, combination of the IV transplantation of allogenic mouse BMMNCs and intermittent physical and cognitive training via repeated behavioral testing was studied in a mouse permanent MCAO model. 37 The combined therapy was found to improve long‐term cognitive, sensorimotor, and motor outcomes 8 and 12 weeks after the onset of stroke when compared to training alone. Plausible cell‐induced or combination effect could not be established, however, as there were no vehicle control and cell treatment‐only groups in the study. Nevertheless, genes related to energy metabolism were upregulated in both ipsilateral and contralateral hemispheres after the combined treatment. 38 Altogether, IV transplantation of MSCs combined with forced physical training or EE can potentially exert an additive or synergistic therapeutic effect on functional recovery after stroke. However, further studies are still required to investigate the effects of several rehabilitation modalities on various cell types delivered via different systemic routes, including IA transplantation. A better understanding of the more complex therapeutic mechanisms underlying the localized and remote effects of systemic transplantation is also needed. HOW TO MAXIMIZE THERAPEUTIC EFFECT IN A COMBINED CELL‐BASED THERAPY AND REHABILITATION FOR STROKE RECOVERY To maximize treatment effect in a cell‐based therapy for stroke recovery, enhanced cell survival, recruitment, or replacement must be met with effective structural integration and functional connectivity in the host brain environment. Taking their extended stand‐alone therapeutic window into account, the combination of cell‐based therapy and rehabilitation is a promising and feasible approach to accomplish the ultimate goal of facilitating network reorganization and other brain repair mechanisms to result in a complete stroke recovery. The behavioral recovery induced by rehabilitation in stroke patients and models may be through compensation or substitution processes in healthy neurons or partly spared networks rather than via “true” neural repair and restitution of lost circuits. 57 However, there is a massive potential in the ability of transplanted cells or their biomolecules to communicate with surrounding cortical and remote brain regions to bridge this gap in regeneration and repair. Rehabilitation can then be used to train newly formed graft‐host circuitries to aid in structural and functional repair processes, and further assist in functional restitution by acting as a reinforcer of intrinsic and treatment‐induced newly formed connections. 12 Understanding how stroke recovery is mediated by brain plasticity and learning mechanisms, which are pivotal to many rehabilitation strategies, is crucial in the advancement of this novel combination approach for stroke recovery. It has recently been proposed in the preclinical stroke setting that cell‐based therapy may enhance endogenous plasticity and direct structural brain reorganization to follow rehabilitation‐induced reinforcement of newly formed connections. 14 , 16 This means that environmental factors and physical training can be used to stimulate the graft site to enhance neuroplasticity and promote the endogenous repair mechanisms that favor exogenous cell survival and integration. 58 , 59 An overall synergistic regenerative microenvironment or much enhanced brain tissue repair has already been achieved with this combination strategy in rodent stroke models. Still, the desired goal of utilizing experimental rehabilitation to facilitate repair processes and the newly formed circuitries to result in maximized behavioral outcomes remains unmet. This suggests that the most significant therapeutic effects on behavioral recovery after stroke may only be achieved if the immature transplanted cells are adequately trained to form effective graft–host structural integration and functional connectivity in the brain. 12 It is not clear, though, whether the plasticity‐related repair and learning mechanisms, anatomical substrates (lost or spared), and functional connectivity pathways underlying spontaneous recovery as well as cell‐ and/or rehabilitation‐induced behavioral improvement after stroke are the same. Moreover, how to therapeutically target the well‐orchestrated and highly interwoven regenerative and reparative mechanisms in such a way that beneficial effects are potentiated or maximized still presents a significant challenge. Nevertheless, voluntary exercise, forced physical training, and EE confer distinct reparative effects that may synergistically or maximally interact with cell‐based therapy if certain guidelines with regard to delivery route, timing, sequence, dosage, and intensity are optimal. 14 Otherwise, the complicated underlying mechanisms could yield overlapping or antagonistic interactions and prohibit the maximization of therapeutic efficacy, or even worsen outcomes, if the stand‐alone therapies are not effectively combined. 13 The mechanisms underlying the endogenous recruitment of NSPCs have ushered in a new avenue for understanding spontaneous recovery after stroke. 60 The endogenous NSPCs promote neuroprotection early after stroke and mediate proliferation when the rodents are still in the initial stage of recovery, which can be further enhanced by cell transplantation. To achieve a better stand‐alone therapeutic effect on functional recovery, immature exogenous cells may be primed or optimized before transplantation. Promising pre‐transplantation optimization strategies include cell encapsulation or scaffolding with biomaterials and gene transfection of cell adhesion molecules, cytokines, chemokines, or growth factors. Cell preconditioning with hypoxia, inflammatory mediators, or photo‐ and optochemogenetics is also emerging. Transplantation of cell aggregates or multilayered cell sheets as well as cell‐derived exosomes, mitochondria, or secretome‐conditioned media also holds great promise. 6 , 7 , 13 As soon as the stroke rodents start displaying functional recovery, rehabilitation should be implemented at an appropriate time to increase neurogenesis and neuronal integration into the host tissue. 6 , 12 It is imperative that the increase in endogenous NSPCs after cell‐based therapy must be followed up with not only an increased training but also an appropriate rehabilitation modality that would allow the immature endogenous and exogenous cells to properly mature and effectively integrate with the host tissue. Voluntary exercise, forced physical training, and EE have all been shown to improve motor function by increasing BDNF and neurogenesis, which is one of the main driving forces underlying stroke recovery. 24 , 25 , 43 , 61 However, behavioral task specificity and integration need to be more widely incorporated in experimental rehabilitation strategies to achieve better functional outcomes in stroke animals. 39 Motor recovery after stroke is a complex process involving functional restitution, compensation, or substitution in neural tissue and the ability to perform voluntary movements. Therefore, rehabilitation of neuromuscular functions is targeted at specific sensation‐movement neuronal circuits. 62 In stroke patients, task‐specific training improves functional motor recovery by inducing an adaptive neural plasticity characterized by synaptogenesis and cortical reorganization in the targeted brain regions. 63 Accordingly, the incorporation of a skilled training modality into voluntary exercise, forced physical training, or EE has been proposed for stroke animals, though the maximal therapeutic dose needs to be verified. In fact, the combination of EE and task‐specific skilled training synergistically improves skilled forelimb‐reaching ability in stroke rats. 64 , 65 EE is unstandardized in the preclinical stroke setting, though, and is also typified by a huge variability in the configurations of physical, sensorimotor, cognitive, or social enrichment. Interestingly, a three‐phase, novelty/complexity‐escalating EE paradigm corresponding to the acute, subacute, and chronic phases of stroke has been proposed as a strategy to limit variability and maximize therapeutic dosage. 24 This EE paradigm ameliorates learning impairments in stroke rats by intensifying NSPC proliferation in the SVZ, promoting neuronal survival in the striatum and cortex, and boosting repair mechanisms in the brain. Importantly, modalities that effectively minimize the variability in overall stimulation or animal activity as well as maximize the intensity or the task‐specific component in experimental rehabilitation strategies need to be verified. There is a defined and critical plasticity window after stroke during which neuroplasticity can be targeted for better functional outcomes, but the optimal timing of each therapy in a combination approach remains undecided. 14 The neurorepair and recovery mechanisms in animal stroke models are widely believed to be at the peak of their plasticity between postoperative Days 1 and 7, perhaps corresponding to the optimal therapeutic window for the transplantation of cells or their biomolecules. 6 Voluntary exercise or forced physical training, if implemented too early, results in worsen functional outcomes, and should be further explored in combination with cell‐based therapy. 55 On the other hand, EE is safe and effective when implemented at early or late time points after stroke. 24 , 66 EE also possesses the added advantage of potentially acting as both a priming and a consolidation therapy. 14 In summary, therapeutic effect can be maximized in experimental stroke animals if (i) optimized exogenous cells or their biomolecules are transplanted early after stroke to further promote cell potency or replacement as well as better enhance and prime repair‐related neuroplasticity and (ii) the timely and optimal dosage of an intensive, task‐specific, and multimodal rehabilitation strategy is used to train immature endogenous and exogenous cells to form consolidated graft–host structural integration and functional connectivity towards a more effective brain repair. In turn, this would foster neural restitution relative to compensation or substitution processes and therefore bridge the gap in neuroregeneration and neurorepair in functional recovery after stroke. It could also guide the optimization of this combination approach and facilitate its advancement in the stroke recovery field. Of note, the combination of cell‐based therapy and rehabilitation for better stroke outcomes is an emerging concept with a lack of strong experimental evidence to support the conclusions on its therapeutic efficacy. Notwithstanding the challenges associated with its feasibility in the preclinical setting, such as time and budgetary constraints as well as a complicated study design, the potential of this combination approach for the maximization of stroke recovery suggests that further studies to verify or establish its promise are warranted. FUTURE PROSPECTS AND CONCLUSIONS Emerging experimental evidence shows that the combination of cell‐based therapy and rehabilitative training is a feasible option to further promote or maximize stroke recovery. Most stroke survivors actually receive some form of rehabilitation, although how much, when, and exactly what vary a lot. 67 The hypothesis that cell‐based therapy opens brain plasticity window that rehabilitative training modulates is still valid and should guide future clinical cell‐based research. The major translational challenge is how to increase therapeutic contrast to differentiate long‐term stand‐alone and combination treatment effects. Choosing a proper stroke model and outcomes according to STEPS 4 15 and SRRR 68 guidelines could be the first step in this optimization process. Small infarcts in the critical cortical structures produce long‐lasting impairments, which can be assessed by robust and sensitive behavioral outcome measures. 15 Selection of rehabilitation treatment for rodents should be based on relevant human situations such as described for upper limb recovery trial. 67 The optimization of the treatment sequence is still needed, that is, whether the application of the priming and consolidation therapies should be concurrent or sequential. 14 , 16 Large‐scale, high‐quality, and clinically aligned multicenter preclinical trials to better understand or verify the mechanisms underlying the stand‐alone, add‐on, and combined effects may be essential, however, before a maximally effective combination therapy can be planned or a successful clinical translation can be achieved. AUTHOR CONTRIBUTIONS Abdulhameed Bakreen: Conceptualization (equal); data curation (lead); visualization (lead); writing—original draft (lead); writing—review and editing (equal). Jukka Jolkkonen: Conceptualization (equal); supervision (lead); writing—review and editing (equal). CONFLICT OF INTEREST STATEMENT Jukka Jolkkonen is the Editorial Board member of Neuroprotection. He is therefore excluded from the peer‐review process and all editorial decisions related to the publication of this manuscript. The remaining authors declared that they have no other conflicts of interest with reference to this manuscript. ETHICS STATEMENT The authors have nothing to report. ACKNOWLEDGMENTS The authors acknowledge the important support of the RESSTORE Steering Committee members: Olivier Detante, Exuperio Díez‐Tejedor, Blanca Fuentes, Marc Hommel, Assia Jaillard, Robert Mikulik, Anaick Moisan, Francisco Moniche, Joan Montaner, Alejandro Bustamante, Keith Muir, Heikki Numminen, David Koubi, Claire Bollart, and Tuija Keinonen. This work was supported by the RESSTORE translational project (www.resstore.eu) funded by the European Commission under the Horizon 2020 program (Grant agreement ID: 681044). Bakreen A, Jolkkonen J. The potential of a combined cell‐based therapy and rehabilitation approach for stroke recovery. Neuroprotection. 2023;1:117‐129. 10.1002/nep3.27 Managing Editor: Ningning Wang DATA AVAILABILITY STATEMENT All relevant data are within the paper. REFERENCES Associated Data Data Availability Statement All relevant data are within the paper.